st: Re: statalist-digest V4 #3872

[Christopher Baum <kit.baum@bc.edu>]

<>
On Aug 6, 2010, at 2:33 AM, Burak wrote:

> I am trying to estimate an hybrid New Keynesian Phillips curve. Currently I
> am using ivreg2 with the following command
> 
> ivreg2 gnd dum11 dum22 dum33 L.gnd mc (F.gnd = L.gnd L2.gnd L3.gnd L4.gnd
> L.cpi L2.cpi L3.cpi L4.cpi L.rint L2.rint L3.rint L4.rint L.bpfilt L2.bpfilt
> L3.bpfilt L4.bpfilt L.mc L2.mc L3.mc L4.mc), gmm2s robust bw(auto)
> 
> what I want to do is to preimpose a restriction such that the coefficients
> of L.gnd and F.gnd sum up to one, and then estimate this. Is there a way to
> do this?



ivreg2 D.gnd dum11 dum22 dum33 mc (FLgnd = L(1/4).gnd etc.

where FLgnd = F.gnd - L.gnd

Works the same way as imposing CRTS algebraically on a two-factor production fn. The
coefficient on FLgnd is the coefficient of F.gnd in the original eqn. lincom can calculate the 
point and interval estimate of its complement (the coefficient of L.gnd in the original eqn.)

Kit

Kit Baum   |   Boston College Economics & DIW Berlin   |   http://ideas.repec.org/e/pba1.html
                              An Introduction to Stata Programming  |   http://www.stata-press.com/books/isp.html
   An Introduction to Modern Econometrics Using Stata  |   http://www.stata-press.com/books/imeus.html


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/